基于微多普勒的弹道多目标分离方法

摘要

弹道多目标分离对弹道导弹早期预警和识别具有重大意义。目前弹道目标微动提取与识别的研究都是基于“单目标多散射点”和“多目标单散射点”的假设，而对实际中弹道“多目标多散射点”的情况缺乏考虑。针对这一问题，首先分析了进动锥体弹头和摆动锥体诱饵的微多普勒形式，得到了同一目标不同散射点的微多普勒具有相同的周期性。对于多目标分离问题，首先利用Radon变换估计平动参数实现了多目标平动补偿；之后通过分析多目标时频图循环平稳性，发现弹道多目标分离本质上是多个二维一阶循环平稳（first-order cyclostation-ary，FOCS）分量的分离问题；其次，提出了一种基于二维FOCS处理的多目标分离方法；最后，通过仿真验证了该方法的有效性和在强噪声下的稳定性。%Multi-ballistic targets resolution is significant for ballistic missile early warning and recognition. At present,the study on the micro-motion ballistic missile is confined in the assumption of“a single target with multi-scattering centers”and “multi-targets with a single scattering center”.To solve this problem,micro-Doppler of precessing missile and swinging decoy are firstly derived.And it proves that the micro-Doppler with the same target and different scattering centers has the same cycle.Then radon transform is applied to compen-sate the translation.After analyzing cyclostationarity of multi-targets time-frequency plane,we regard multi-bal-listic targets resolution as a problem to resolve a signal composed of several two-dimensional first-order cyclostationary (FOCS).A method based on the two-dimensional FOCS is proposed to resolve multi-targets.Finally,simulations are given to validate the effectiveness and the robustness in high noise environment of this method.